read more - Current Science

CURRENT SCIENCE
Volume 108
Number 2
25 January 2015
GUEST EDITORIAL
Bibliometrics – problems and promises
‘Bibliometrics has a problem. In fact, the field – which
tracks scholarly impact in everything from journal papers
to citations, data sets and tweets – has a lot of problems.
Its indicators need better interpretation, its tools are
widely misused, its relevance is often questioned and its
analyses lack timeliness.’
– from the review by Jonathan Adams of the book
Beyond Bibliometrics: Harnessing Multidimensional
Indicators of Scholarly Impact
(eds Cronin, B. and Cassidy, R.), MIT Press, 2014;
which appeared in Nature, 2014, 510, 470–471.
Long-time readers of Current Science will be familiar
with its somewhat ambivalent attitude to the discipline
that now goes by several names: bibliometrics, scientometrics, informetrics, evaluative bibliometrics, research
evaluation, altmetrics, etc. This is well reflected in the
dozen editorials or so that have appeared in the journal
over the years from 1998 to 2013.
There is a saying that ‘if you can’t measure it, you
can’t manage it’. Bibliometrics became prominent
because of the need to manage the huge investments
that were going into the science and technology (S&T)
sectors and into research and development activities in
particular.
As early as 1939, J. D. Bernal made an attempt to
measure the amount of scientific activity in a country and
relate it to the economic investments made. In The Social
Function of Science (1939), Bernal estimated the money
devoted to science in the United Kingdom using existing
sources of data: government budgets, industrial data
(from the Association of Scientific Workers) and University Grants Committee reports. He was also the first to
propose an approach that became the main indicator of
S&T: gross expenditures on research and development
(GERD) as a percentage of GDP. He compared the
investment of UK at that time (0.1%) with that of the
United States (0.6%) and USSR (0.8%), and suggested
that Britain should devote 0.5–1.0% of its national
income to research. Since then, research evaluation at the
country and regional levels has progressed rapidly and
there are now exercises carried out at regular intervals in
USA, European Union, OECD, UNESCO, Japan, China,
CURRENT SCIENCE, VOL. 108, NO. 2, 25 JANUARY 2015
etc. It is important to note here that India’s GERD to
GDP ratio has never crossed 1%!
Science is a socio-cultural activity that is highly disciplined and easily quantifiable. The complete chain can be
thought of as follows: input  processes  output 
outcome  impact. The last two steps are the most difficult to measure and usually involve cognitive skills that
go beyond bibliometrics. Bibliometrics helps one measure the output of science relatively easily in terms of articles published and citations to these, etc. Inputs are
mainly those of the financial and human resources invested in S&T activity. The financial resources invested
in research are used to calculate the GERD, and the
human resources devoted to these activities (FTER – full
time equivalent researcher) are usually computed as a
fraction of the workforce or the population. The US science adviser, J. R. Steelman pointed out in 1947 that ‘The
ceiling on research and development activities is fixed by
the availability of trained personnel, rather than by the
amounts of money available. The limiting resource at
the moment is manpower’.
Although the discipline has origins that go back to the
early years of the last century, it became data-driven and
evidence-based exactly 50 years ago – when the Science
Citation Index was launched in 1964, a brain-child of the
legendary Eugene Garfield. It came with a lot of promises
but the problems arose because, in the words of Jonathan
Adams, ‘the non-experts – such as reviewers, hiring
committees and grant panels [use] and often misuse
bibliometrics to make decisions.’
Let us first start with an example that is close to us all.
Without such measurements we would have no idea, except from anecdotal evidence, where we stand as a nation
in the S&T space. Derek John de Solla Price, one of the
pioneers of the field, showed around the mid-sixties that
the size of the scientific enterprise can be measured
crudely by the number of first authors who publish papers
and he established a cursory relationship that this was
approximately proportional to the economic size of the
country as measured by the Gross National Product. So
around 1967, India ranked ninth in economic size (just
behind Canada) and eighth in scientific size (again just
behind Canada). That is, our science stayed ahead of our
147
GUEST EDITORIAL
economy. Today, nearly half a century later, while we are
the eleventh largest economy in the world in nominal
GDP terms, we rank in bibliometric terms, around the
sixteenth position in a simple quantity measure (scientific
output), and around the twenty-eighth position in quality
terms (measured by scientific output per million dollars
of GDP). That is, our science is not even as good as our
economy says it should be.
Evidence of this slow and steady decline was already
known to our practitioners of bibliometrics and has been
recorded in the very own pages of Current Science. From
the first half (1980–84) to the second half (1985–89) of
the decade of the eighties, India’s total contribution to the
world’s publication output (as measured by the Science
Citation Index database) dropped by 17.8%, while the
world output increased by 9.7% (ref. 1). A more detailed
study2 covering the two decades from 1980 to 2000,
showed that while Chinese science rose by a factor of 23
(from 924 papers in 1980 to 22,061 papers in 2000),
Indian science actually slowed down (from 14,983 papers
in 1980 to 12,127 papers in 2000). This is bibliometrics
at its best, when applied at the macro-level. It is not clear
whether our science administrators and bureaucrats learned
any lessons from such bibliometric investigations.
Bibliometric studies at the meso-level (institutions,
agencies, disciplines, journals, etc.) can also give useful
insights into how funding should be targeted and channellized. In this manner, it serves the needs of research
evaluation and management well. At the micro-level, for
assessing individual scientists for awards and promotions,
or for project proposals for grants, it has to be used with
caution. Here, citation-based measures should complement expert peer-review rather than supplant it.
It is at the level of science and understanding that bibliometrics has gone far beyond its earliest promises. Garfield’s revolutionary idea that science can be tracked by
an association of ideas, and the Francis Narin and Gabriel
Pinski proposal to use recursive iteration and weighting
to separate prestige from mere popularity led to the
merging of social networking and graph theoretical tools
that inform much of how the web-ranking algorithms
such as the one found in Google’s PageRank are now
148
implemented and used. Such tools have now come back
(recursively?) into bibliometrics to produce indicators
like Eigenfactor, Article Influence and the Scopus Journal
Ranking scheme.
This is not to say that the temptation to use bibliometrics carelessly and thoughtlessly can be easily checked.
It will be useful to remember Albert Einstein’s caveat –
‘Everything that can be counted does not necessarily
count; everything that counts cannot necessarily be
counted.’
One of the gravest dangers to the way science is practised is the pervasive, and sometimes perverse, influence
that Hirsch’s h-index has had on the collective imagination. Bibliometrics can now be divided into a pre-Hirsch
and a post-Hirsch phase. More dangerously, science is
getting increasingly polarized into groups that do science
for the pleasure of unravelling the hidden truths of the
universe and those who pursue science simply to enhance
their h-index. I recall the words of a famous bibliometrics
expert Tibor Braun, the founder editor of the journal
Scientometrics, when I met him a few years ago: ‘Nowadays, everyone wants to know the h-index of his enemy’.
It is not rare to find scientists who now consult almost on
a daily basis databases such as the Thomson Reuters Web
of Science, Elsevier’s Scopus and Google Scholar to see
what their h-index is! Nor is it unusual to find research
managers and funding agencies who demand that the
h-indices be provided as part of the curriculum vitae.
Indeed, if this ‘stranglehold of the h-index and the impact
factor is loosened science might again be fun’3.
1. Prathap, G., Curr. Sci., 1995, 68, 983–984.
2. Arunachalam, S., Curr. Sci., 2002, 83, 107–108.
3. Balaram, P., Curr. Sci., 2009, 96, 1289–1290.
Gangan Prathap
CSIR-National Institute for Interdisciplinary Science
and Technology,
Thiruvananthapuram 695 019, India
e-mail: gp@niist.res.in
CURRENT SCIENCE, VOL. 108, NO. 2, 25 JANUARY 2015